Ir para o conteúdo
ModelListadosExterno
Baixar

Manga Ocr Finetuned

ONNX export of JustANormalTinkerer/manga-ocr-finetuned — a finetuned manga-OCR with a distilled 192-dim ViT encoder (the upstream ships PyTorch only).

Fontehuggingface.co/NorwayFish/manga-ocr-finetuned/tree/1613e9822417ce836df8a49571f97e73fa5f8486
TipoCLIP visionModelo baseJustANormalTinkerer/manga-ocr-finetunedVersãoBnota

Fonte do modelo

Tipo
CLIP vision
Tamanho do arquivo
113 MB
Modelo base
JustANormalTinkerer/manga-ocr-finetuned
Versão
vdecoder_model
Atualizado
30 de out. de 2026
Downloads
4
Fonte
Hugging Face

Descrição da fonte

manga-ocr-finetuned (ONNX)

ONNX export of JustANormalTinkerer/manga-ocr-finetuned — a finetuned manga-OCR with a distilled 192-dim ViT encoder (the upstream ships PyTorch only).

  • Weights: encoder_model.onnx (22 MB, 192-dim encoder) + decoder_model.onnx (118 MB, non-merged), exported with optimum. I/O: encoder pixel_values → last_hidden_state; decoder input_ids + encoder_hidden_states → logits.
  • Tokenizer: the manga-ocr 6144-token WordPiece tokenizer.json, verified vocab-identical to this model's lineage.

Repackaged for single-repo, self-consistent ONNX inference.

Fontes

1 fonte
Hugging FacePrimárioVersão 1613e9822417ce836df8a49571f97e73fa5f8486
huggingface.co/NorwayFish/manga-ocr-finetuned/tree/1613e9822417ce836df8a49571f97e73fa5f8486

Verificado 30 de out.

Download files

Repository files

File inventory only; completeness is unverified.

Artefatos de modelo

1 artefato

decoder_model.onnx

onnx · 113 MB · SHA-256 4258a0daf4fd…da8a · Hugging Face

Download file

Trechos de fonte

3 trechos
Readme14 de ago.huggingface.co/NorwayFish/manga-ocr-finetuned

--- license: apache-2.0 base_model: JustANormalTinkerer/manga-ocr-finetuned language: - ja pipeline_tag: image-to-text tags: - onnx - vision-encoder-decoder - image-to-text - manga-ocr --- # manga-ocr-finetuned (ONNX) ONNX export of [`JustANormalTinkerer/manga-ocr-finetuned`](https://huggingface.co/JustANormalTinkerer/manga-ocr-finetuned) — a finetuned manga-OCR with a distilled **192-dim** ViT encoder (the upstream ships PyTorch only). - **Weights**: `encoder_model.onnx` (22 MB, 192-dim encoder) + `decoder_model.onnx` (118 MB, non-merged), exported with `optimum`. I/O: encoder `pixel_values` → `last_hidden_state`; decoder `input_ids` + `encoder_hidden_states` → `logits`. - **Tokenizer**: the manga-ocr 6144-token WordPiece `tokenizer.json`, verified vocab-identical to this model's lineage. Repackaged for single-repo, self-consistent ONNX inference.

Note30 de out.huggingface.co/NorwayFish/manga-ocr-finetuned

Source context: 4 downloads · 0 likes · Pipeline image-to-text · Repo NorwayFish/manga-ocr-finetuned

Note14 de ago.huggingface.co/NorwayFish/manga-ocr-finetuned

Source context: 10 downloads · 0 likes · Pipeline image-to-text · Repo NorwayFish/manga-ocr-finetuned